Applications and Comparison of Continuous Wavelet Transforms on Analysis of A-wave Impulse Noise

Abstract

Noise induced hearing loss (NIHL) is a serious occupational related health problem worldwide. The A-wave impulse noise could cause severe hearing loss, and characteristics of such kind of impulse noise in the joint time-frequency (T-F) domain are critical for evaluation of auditory hazard level. This study focuses on the analysis of A-wave impulse noise in the T-F domain using continual wavelet transforms. Three different wavelets, referring to Morlet, Mexican hat, and Meyer wavelets, were investigated and compared based on theoretical analysis and applications to experimental generated A-wave impulse noise signals. The underlying theory of continuous wavelet transform was given and the temporal and spectral resolutions were theoretically analyzed. The main results showed that the Mexican hat wavelet demonstrated significant advantages over the Morlet and Meyer wavelets for the characterization and analysis of the A-wave impulse noise. The results of this study provide useful information for applying wavelet transform on signal processing of the A-wave impulse noise.Keywords: continuous wavelet transform, impulse noise signal processing, time-frequency domain, temporal and spectral resolutions, noise induced hearing loss, A-wave impulse noise.

A wavelet-based hybrid islanding detection system applied for distributed generators interconnected to AC microgrids

“…The wavelet function ( ) a b , has dilation a ( ) proportional to translation b ( ) with its time support centered at dilation [47]. Therefore, the CWT can decompose a signal into the time-frequency representation and guarantee the temporal and spectral resolutions in the frequency range [48,49]. The mother wavelet ( ) is a window function that builds the wavelet transform and can be written compactly from normalization mother wavelets as follows [50]:…”

“…. The Morlet mother wavelet is well-suited for non-stationary signals analysis providing the detection of their irregularity characteristics through its localized filters in the time-frequency domain [49,51]. For that reason, it has been commonly used in different applications to supervise power system disturbances [34,[51][52][53][54], identify oscillation modes [55], damp ratios and natural frequencies [56], estimate frequency and phasor [57], diagnose faults [52,58,59], and estimate grid impedance [46].…”

Archives of Acoustics

“…For time-frequency analysis, there are some other approaches including the short-time Fourier transform (STFT), wavelet transform (WT), Wigner-Ville distribution (WVD), and the Hilbert-Huang transform (HHT), which are usually employed for feature extraction of non-stationary signals (Błazejewski et al, 2014;Huang et al, 2015). The continuous wavelet transform (CWT) is commonly used for data analysis, while the discrete wavelet transform (DWT) is applied for image compression and pattern recognition (Mallat, 2009;Qin, Sun, 2015). The Wigner-Ville distribution is a popular approach thanks to its good time-frequency resolution, however, generates results with coarser granularity than those of the wavelet transform methods (Xing et al, 2016).…”